Background
Maternal mortality ratio in Nigeria is one of the highest in the world. Near misses occur in larger numbers than maternal deaths hence they allow for a more comprehensive analysis of risk factors and determinants as well as outcomes of life-threatening complications in pregnancy. The study determined the incidence, characteristics, determinants and perinatal outcomes of near misses in a tertiary hospital in South-west Nigeria.

Methods
A prospective case control study was conducted at the maternity units of the Obafemi Awolowo University Teaching Hospitals Complex, Ile-Ife Nigeria between July 2006 and July 2007. Near miss cases were defined based on validated disease-specific criteria which included severe haemorrhage, hypertensive disorders in pregnancy, prolonged obstructed labour, infection and severe anemia. Four unmatched controls of pregnant women were selected for every near miss case. Three categories of risk factors (background, proximate, clinical) which derived from a conceptual framework were examined. The perinatal outcomes were also assessed. Bi-variate logistic regressions were used for multivariate analysis of determinants and perinatal outcomes of near miss.

Results
The incidence of near miss was 12%. Severe haemorrhage (41.3%), hypertensive disorders in pregnancy (37.3%), prolonged obstructed labour (23%), septicaemia (18.6%) and severe anaemia (14.6%) were the direct causes of near miss. The significant risk factors with their odds ratio and 95% confidence intervals were: chronic hypertension [OR=6.85; 95% CI: (1.96 – 23.93)] having experienced a phase one delay [OR=2.07; 95% CI (1.03 – 4.17)], Emergency caesarian section [OR=3.72; 95% CI: (0.93 – 14.9)], assisted vaginal delivery [OR=2.55; 95% CI: (1.34 – 4.83)]. The protective factors included antenatal care attendance at tertiary facility [OR=0.19; 95% CI: (0.09 – 0.37)], knowledge of pregnancy complications [OR=0.47; 95% CI (0.24 – 0.94)]. Stillbirth [OR=5.4; 95% CI (2.17 – 13.4)] was the most significant adverse perinatal outcomes associated with near miss event.

Conclusions
The analysis of near misses has evolved as a useful tool in the investigation of maternal health especially in life-threatening situations. The significant risk factors identified in this study are amenable to appropriate public health and medical interventions. Adverse perinatal outcomes are clearly attributable to near miss events. Therefore the findings should contribute to Nigeria’s effort to achieving MDG 4 and 5.